The age-associated increase in cancer may be due in part to a global decrease in immune function. It has been thought that this loss of immune function was intrinsic to the aged lymphoid cells (i.e. aged T cells are defective). However, recent data from our laboratory utilizing a tumor model indicate that if the immune response is stimulated in the presence of co-stimulatory molecules, naive aged CDS1 T cells are able to undergo primary activation with the same degree of vigor as that of young CD8+ T cells, including the development of long-lasting memory responses. These results show that old mice do possess a functional T cell repertoire that can be activated and expanded if there is sufficient costimulation. This data led us to propose the hypothesis that the aged T cell repertoire can be exploited for the induction of tumor immunity. However, to effectively stimulate an antitumor response it is critical to identify and optimize vaccination protocol(s) that will result in the generation of primary and long-lasting memory responses. This proposal will evaluate vaccination strategies in which each system uses a different mode of action to target and activate antigen-presenting cells (APCs) to increase antigen presentation. We will utilize three different "adjuvants/vaccines" such as Adenoviruses, flagellin-fusion proteins or complement-derived-fusion proteins in combination with costimulatory molecules such as anti-OX40 or anti-4-lBB to optimize antitumor immune responses in old hosts. To investigate the efficacy and optimize these vaccines, we will use two different tumor models: The first model utilizes the Enhanced Green Fluorescent protein (EGFP) as a surrogate tumor antigen to evaluate immune responses to a potent antigen in which it is not a self-antigen. The second model utilizes the Her-2/neu protein that is a more clinical relevant tumor antigen in which it is a self-antigen. In aim 1 we will compare in vitro CDS T cell responses of young and old mice after vaccination with the Adenoviruses, flagellin-fusion proteins or complement-derived-fusion proteins expressing the EGFP or Her-2/neu gene/protein. We will compare different doses and routes and follow the persistence and effectiveness of the primary and memory responses over time. These experiments will allow us to define which protocol induces the strongest and persistent immune response in old mice. Aim 2 will test the most efficient vaccination protocol (defined in Aim 1) to induce protective immunity or for the treatment of established tumors. Based on the accumulative data we have generated with our animal tumor models, we predict that immunotherapy treatment could substantially delay the tumor growth but it will not be sufficient for the complete elimination of tumors in the aged. In Aim 3 we will test the hypothesis that targeted immunotherapy in combination with antiangiogenic therapy can be synergistic rendering a more efficient mechanism for the complete elimination of tumor in old animals. The information gained from these studies could have important clinical implications for the development of vaccination strategies in the elderly.